Pathological, excessive neovascularization is the primary cause of blindness in proliferative diabetic retinopathy (PDR). In general, vascular remodeling and angiogenesis is required for the progression of diabetes and many other life-threatening diseases. Our long-term research goal is to elucidate the complex, multivariate regulatory mechanisms governing pathological angiogenesis and vascular remodeling by analyzing change in vascular pattern as a read-out of molecular regulation. In preliminary studies using our novel region-based fractal analysis, vascular density decreased during nonproliferative diabetic retinopathy (NPDR) prior to the excessive neovascularization that clinically defines progression to PDR. We therefore propose to first validate the hypothesis that (1) vascular density decreases during NPDR prior to the pathological neovascularization of PDR, and further test that (2) the initial decrease in vascular density during NPDR may be reversible with promising, new anti-angiogenic therapeutics currently undergoing clinical trials for treatment of diabetic retinopathy. To test our hypothesis, the following questions posed in Specific Aims will be answered by region-based fractal analysis: (1) How do blood vessels remodel during early-stage, non-proliferative diabetic retinopathy (NPDR) prior to the pathological neovascularization that defines late-stage, proliferative diabetic retinopathy (PDR)? Does vascular density continue to decrease during progression from mild to severe NPDR? Prospective clinical studies of the NPDR human retina will be performed. (2) Can the molecular regulation of vascular remodeling during NPDR by promising steroids and other anti-VEGF inhibitors be deduced and quantified by region-based fractal analysis? Effects of the steriod triamcinolone acetonide will be quantified in the NPDR human retina and in another 2D, optically accessible tissue of similar vascular morphology, the quail chorioallantoic membrane (CAM). Currently it is not known how blood vessels remodel during NPDR, although capillary dropout during NPDR is well established. To study how blood vessels remodel in the 2D, optically accessible NPDR human retina, we are applying fractal techniques first developed in the CAM. The CAM continues to serve as a convenient testbed for ongoing development of fractal methods and for examination of effects of human regulators on angiogenesis and vascular remodeling.